代码拉取完成,页面将自动刷新
同步操作将从 qzl66/Qwen-Agent 强制同步,此操作会覆盖自 Fork 仓库以来所做的任何修改,且无法恢复!!!
确定后同步将在后台操作,完成时将刷新页面,请耐心等待。
import re
from setuptools import find_packages, setup
def get_version() -> str:
with open('qwen_agent/__init__.py', encoding='utf-8') as f:
version = re.search(
r'^__version__\s*=\s*[\'"]([^\'"]*)[\'"]',
f.read(),
re.MULTILINE,
).group(1)
return version
def read_description() -> str:
with open('README.md', 'r', encoding='UTF-8') as f:
long_description = f.read()
return long_description
# To update the package at PyPI:
# ```bash
# python setup.py sdist bdist_wheel
# twine upload dist/*
# ```
setup(
name='qwen-agent',
version=get_version(),
author='Qwen Team',
description='Qwen-Agent: Enhancing LLMs with Agent Workflows, RAG, Function Calling, and Code Interpreter.',
long_description=read_description(),
long_description_content_type='text/markdown',
keywords=['LLM', 'Agent', 'Function Calling', 'RAG', 'Code Interpreter'],
packages=find_packages(exclude=['examples', 'examples.*', 'qwen_server', 'qwen_server.*']),
package_data={
'qwen_agent': [
'utils/qwen.tiktoken', 'tools/resource/*.ttf', 'tools/resource/*.py', 'gui/assets/*.css',
'gui/assets/*.jpeg'
],
},
# Minimal dependencies for Function Calling:
install_requires=[
'dashscope>=1.11.0',
'eval_type_backport',
'json5',
'jsonlines',
'openai',
'pydantic>=2.3.0',
'requests',
'tiktoken',
],
extras_require={
# Extra dependencies for RAG:
'rag': [
'charset-normalizer',
'rank_bm25',
'jieba',
'snowballstemmer',
'beautifulsoup4',
'pdfminer.six',
'pdfplumber',
'python-docx',
'python-pptx',
'pandas',
'tabulate',
],
# Extra dependencies for Code Interpreter:
'code_interpreter': [
'anyio>=3.7.1',
'fastapi>=0.103.1',
'jupyter>=1.0.0',
'matplotlib',
'numpy',
'pandas',
'pillow',
'seaborn',
'sympy',
'uvicorn>=0.23.2',
],
# Extra dependencies for Gradio-based GUI:
'gui': [
'gradio==4.21.0',
'modelscope-studio>=0.4.0',
],
},
url='https://github.com/QwenLM/Qwen-Agent',
)
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。